Machine Learning of Temporal Relations

By Inderjeet Mani, Marc Verhagen, Ben Wellner, Chong Min Lee, James Pustejovsky. Annual Meeting of the Association for Computational Linguistics, 2006

306 citations recorded by the source corpus.

This paper investigates a machine learning approach for temporally ordering and anchoring events in natural language texts. To address data sparseness, we used temporal reasoning as an over-sampling method to dramatically expand the amount of training data, resulting in predictive accuracy on link labeling as high as 93% using a Maximum Entropy classifier on human annotated data. This method compared favorably against a series of increasingly sophisticated baselines involving expansion of rules derived from human intuitions.

Subfields: Artificial Intelligence, Computer Networks and Communications, Information Systems

View Paper: https://www.semanticscholar.org/paper/fc82788021963ff8e318ffe955829bc68e48943a

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Mohit Bansal
2. Nanyun Peng
3. Dan Roth
4. Yejin Choi
5. Jianfeng Gao
6. Percy Liang
7. Christopher D. Manning
8. Noah A. Smith

In Computer Networks and Communications

1. Dan Roth
2. Qiang Ning
3. Rujun Han
4. Nanyun Peng
5. Nathanael Chambers
6. Hao Wu
7. James Pustejovsky
8. Aram Galstyan

In Information Systems

1. Wei Dang Lu
2. Dan Roth
3. Nanyun Peng
4. Rujun Han
5. Shuaicheng Zhang
6. Lifu Huang
7. Qiang Ning
8. Aram Galstyan

Citations in the corpus

• New Regularized Algorithms for Transductive Learning (2009, background): “Recently, graph based semi-supervised algorithms have achieved considerable attention [2] [7] [11] [14] [17].”
• Learning Script Knowledge with Web Experiments (2010, methodology): “These systems typically employ a method for classifying temporal relations between given event descriptions (Chambers et al., 2007; Cham-bers and Jurafsky, 2008a; Mani et al., 2006).”
• Joint Reasoning for Temporal and Causal Relations (2018, background): “Existing work on temporal relation extraction was initiated by (Mani et al., 2006; Chambers et al., 2007; Bethard et al., 2007; Verhagen and Pustejovsky, 2008), Ex 3: Global considerations are needed when making local decisions.”
• Sort Story: Sorting Jumbled Images and Captions into Stories (2016, background): “A number of works (Mani and Schiffman, 2005; Mani et al., 2006; Boguraev and Ando, 2005) learn temporal relations and properties of news events from the dense, expert-annotated TimeBank corpus (Pustejovsky et al., 2003).”
• Extracting Temporal Event Relation with Syntactic-Guided Temporal Graph Transformer (2021, methodology): “Early studies on temporal relation extraction mainly model it as a pairwise classification problem (Mani et al., 2006a; Verhagen et al., 2007; Verhagen and Pustejovsky, 2008; Verhagen et al., 2010; Bethard et al., 2016; MacAvaney et al., 2017) and use statistical machine learning techniques with…”
• Joint Event and Temporal Relation Extraction with Shared Representations and Structured Prediction (2019, background): “Early attempts on temporal relation extraction use local pair-wise classification with hand-engineered features (Mani et al., 2006; Verha-gen et al., 2007; Chambers et al., 2007; Verha-gen and Pustejovsky, 2008).”
• Unsupervised Learning of Narrative Event Chains (2008, methodology): “There are a number of algorithms for determining the temporal relationship between two events (Mani et al., 2006; Lapata and Lascarides, 2006; Cham-bers et al., 2007), many of them trained on the Time-Bank Corpus (Pustejovsky et al., 2003) which codes events and their temporal relationships.”
• Deep Structured Neural Network for Event Temporal Relation Extraction (2019, background): “Early attempts (Mani et al., 2006; Verha-gen et al., 2007; Chambers et al., 2007; Verha-gen and Pustejovsky, 2008) only use local pair-wise classification with hand-engineered features.”
• Joint Constrained Learning for Event-Event Relation Extraction (2020, methodology): “Early effort focused on characterizing event pairs based on various types of semantic and linguistic features, and utilizing statistical learning methods, such as logistic regression (Mani et al., 2006; Verhagen and Pustejovsky, 2008) and SVM (Mirza and Tonelli, 2014), to capture the relations.”
• SemEval-2007 Task 15: TempEval Temporal Relation Identification (2007, methodology): “TimeML and TimeBank have already been used as the basis for automatic time, event and temporal relation annotation tasks in a number of research projects in recent years (Mani et al., 2006; Boguraev et al., forthcoming).”
• Joint Inference for Event Timeline Construction (2012, background): “Recent efforts in building an annotated temporal corpus (Pustejovsky et al., 2003) has pop-ularized the use of machine learning techniques for the task (Mani et al., 2006; Bethard et al., 2007).”
• Open domain event extraction from twitter (2012, background): “before and after) is often required to accurately relate events to temporal expressions [32, 8].”
• Domain Knowledge Empowered Structured Neural Net for End-to-End Event Temporal Relation Extraction (2020, unknown): “Early work on temporal relation extraction use local pair-wise classification with hand-engineered features (Mani et al., 2006; Verhagen et al., 2007; Chambers et al., 2007; Verhagen and Pustejovsky, 2008).”